December 4, 2024
Jeremy Lewin
Labor laws historically have made it challenging for many physicians to unionize because they did not meet the statutory definition of “employee” under the National Labor Relations Act (NLRA). The NLRA states that “[t]he term employee shall include any employee...but shall not include any individual having the status of an independent contractor, or any individual employed as a supervisor.”1 This is a two-part test: the physician must work in an employed arrangement, and they must not be a supervisor.
Whether any particular group of physicians can unionize depends upon whether they are considered employees. Until relatively recently, many physicians were ineligible to unionize because they worked as independent contractors. However, with practice consolidation and other changes in the healthcare industry, a growing percentage of physicians are now employed by hospitals and other corporate entities, satisfying the first part of the NLRA’s two-part test.
Whether a particular employed physician may participate in a union depends upon that individual’s supervisory status. Recent decisions by the National Labor Relations Board (NLRB)—the agency that enforces the NLRA—have clarified that the NLRB does not consider employed physicians with only a limited degree of supervisory authority to be supervisors. For example, a regional office of the NLRB recently found that although a group of hospital physicians had some authority to direct other employees, they were not “supervisors” under the NLRA because their primary purpose was to provide care to patients.
The changing professional landscape for physicians and the NLRB’s recent decisions on supervisory status mean that more physicians are eligible to unionize.
If you ask people in the US whether they support unions, 67% say they do, which is up almost 40% from an all-time low of 48% following the Great Recession.2 Despite this high level of public support, participation in unions is low, at about 10% of wage and salary workers in 2023. That’s about half of what it was in 1983.3 Union representation among physicians appears to be even lower, with estimates generally hovering around 6% to 7%.
However, recent union activity suggests increasing interest in union membership among healthcare workers. In October 2023, more than 500 physicians, nurses, and physician assistants at Allina Health in Minneapolis, Minnesota, voted to unionize by a vote of 325 to 200. The group is represented by Doctors Council Service Employees International Union (SEIU) and is among the largest private sector physician unions in the country. In January 2024, residents, fellows, and interns at Northwestern Medicine in Chicago, Illinois, voted to unionize, citing long hours and low pay. The same month, a group of anesthesiologists in California voted to unionize by a vote of 78 to 25.
Supporters of physician unionization argue that benefits include the ability to collectively bargain for better working conditions, protection from legal action, and the ability to advocate for improved patient care. However, a recent study of US surgical residents found that, although unionized programs may offer better benefits, “unions were not associated with improved burnout, suicidality, job satisfaction, duty hour violations, mistreatment, educational environment, or salary.”4
Other studies have focused on the effects of labor unions on patient outcomes and quality of care. The results have been mixed. A 2020 study of New York nursing homes during the COVID-19 pandemic found that unionized nursing homes had lower mortality rates, better access to personal protective equipment, and stronger infection control measures than their nonunion counterparts.5
A separate study focused on nurse unionization in hospitals concluded that hospitals with successful unionization efforts “experience a decline in the incidence of hospital-acquired illnesses” even when accounting for the effects of strikes on mortality.5 Another study focusing on the general workforce found that unions can result in “increase[d] productivity or product quality through reduced turnover, increased worker effort, and improved worker morale.”5
Other studies have shown ambiguous or negative impacts. A 2010 study focusing on the effects of nurses’ union strikes found that “the average strike increases mortality by more than 18%.”5 Another study “found no impact of unionization on care quality.”5
Notably, many physicians believe striking could run afoul of the Hippocratic Oath’s foundational principle of doing no harm.
While physician unionization opens the theoretical possibility of physician strikes, most strikes in the healthcare system have historically involved residents and nurses. This has traditionally been the case because the “union model permitting strikes is not an option in medicine” either legally, morally, or ethically.6 Notably, many physicians believe striking could run afoul of the Hippocratic Oath’s foundational principle of doing no harm.
Some also worry that the public may not be sympathetic to physician strikes because of the perception of high physician compensation, and the concern that physician strikes may jeopardize public health and safety.
A rare example of physicians striking occurred in 2015. For the first time in its history, the Union of American Physicians and Dentists (UAPD) staged physician strikes at the University of California (UC) system. Approximately 130 physicians staffing student clinics went on strike to protest citing “lack of control over working conditions, stagnant salaries, and lack of compensation for after-hours work.”6
Prior to the strike, UC physicians expressed frustration that after joining the UAPD, they had been negotiating their first contract with the UC system for more than a year. Two strikes occurred with prior notice. One involved all 10 student health clinics and lasted only 1 day. The other focused on the two largest clinics and lasted 4 days. Ultimately, UAPD’s president reported that the union obtained most of what it had requested.
In 2019, resident physicians at the UC San Francisco (UCSF) went on strike to draw attention to continually failing negotiations between their union and the UCSF Medical Center and to express their dissatisfaction with the current grievance procedure. The Committee on Interns and Residents, which represents more than 1,000 UCSF residents, also complained that the medical center underpays and underrepresents residents. To draw attention to these matters, resident physicians, clinical fellows, and interns participated in a 15-minute walk-out, which they called a “unity break.”
In May 2023, more than 150 residents went on strike at Elmhurst Hospital Center in Queens, New York. The strike, which lasted 3 days, was prompted largely by a pay gap between residents at Elmhurst and other affiliated hospitals. The strike ended when the union and the hospital reached a deal that would give Elmhurst residents enhanced pay and benefits, including an 18% pay increase over 3 years.
There is a common misconception that only established unions can represent employees, such as the Steelworkers, Teamsters, or SEIU. That is not accurate. In fact, organizations that wish to represent employees, and employees themselves, can form their own “independent” unions that have no formal ties to an established union. That option has recently become increasingly popular, particularly with younger employees who view established unions as out of touch and not aligned with their viewpoints. Some employees also dislike the for-profit, political nature of established unions.
One example of a newly created union that has received considerable media attention is at Amazon, with subsets of Amazon employees having formed their own union called the Amazon Labor Union. Employees at Starbucks have similarly created the Starbucks Workers United union, although it reportedly is “essentially a front for the Service Employees International Union.”7
As mentioned, organizations other than established unions can form their own independent unions. However, if a medical association were to operate as a union, it would face significant legal and operational challenges. An association’s status as a tax-exempt 501(c)(3) or 501(c)(6) organization requires that it focus its primary activities in service of its tax-exempt purpose, which does not include union activities. Devoting significant association resources to activities unrelated to the association’s tax-exempt purpose could jeopardize an organization’s tax-exempt status.
In light of the risk of losing its tax-exempt status, a medical association would likely need to form a new entity and would apply for tax exemption for that entity under Section 501(c)(5) of the Internal Revenue Code, the section applicable to tax-exempt labor unions. While forming the new entity would help to protect the organization’s tax exemption, the complexities of the union formation process would require a significant investment of time and resources.
Unions employ staff to perform specialized activities that are generally outside the scope of an association’s expertise, including:
The typical geography associated with establishing and operating a union would present further challenges. Because of the legal standards that govern unionization, unions typically represent workers at a single location or multiple locations of a single employer if they are located in close proximity to one another. Very rarely would it be appropriate or permissible for a union to represent employees across a broader geography or employees who work for different employers as a single bargaining unit.
This aspect of unionization would mean the new entity would have to create local offices with their own staff throughout the country. This challenge has contributed to the limited success of independent unions as compared to established unions with a nationwide presence through affiliated “locals” that are responsible for specific geographies.
Whether to form or join a union is a complicated question.
Notably, in 1999, the American Medical Association (AMA) helped form a national labor organization under the NLRA called Physicians for Responsible Negotiation (PRN). The PRN was designed to support the “development and operation of local negotiating units as an option for employed physicians and for resident and fellow physicians.”8 Just 5 years later, “after spending a substantial amount of money on the venture that signed up few physicians, the AMA discontinued financial support for the project.”8
Given the operational challenges outlined earlier, it is not surprising that the AMA ceased funding its union effort after just a few years.
Taking into account the significant resources necessary to support the formation and operation of a union, especially on a nationwide basis, associations may consider partnering with an established union with an existing infrastructure and the experience necessary to manage the complex and cumbersome procedural and operational aspects outlined in this article.
Internal Revenue Service (IRS)-related issues and antitrust risks would present potential challenges. The IRS might object, for example, if it learned that the association gave resources to the union free of charge or at a discount, since the activities these funds would support would not be related to the association’s tax-exempt purpose. These resources would not be limited to cash payments or staff support for the union’s services, and would include assets such as member lists, which the union would surely want in order to access the association’s members.
Partnering with a labor union also may present risk of antitrust violations. Before labor union law was firmly established in the US, many union activities were considered unlawful. Section 1 of the Sherman Act declares that any “contract, combination…or conspiracy in restraint of trade or commerce among the several states” is illegal. The Sherman Act was used effectively in actions against early trade unions.
The US Supreme Court later held that two other Federal acts, the Clayton Act and the Norris-LaGuardia Act, provide a statutory exemption from antitrust laws for certain union activities. More recent statutes, along with favorable judicial interpretation of labor statutes, have resulted in further protections for union activities.
These protections are not absolute, and they do not automatically apply simply because a union is involved. To the contrary, where a union acts with a non labor group to restrain competition in a business market, the activities would constitute an antitrust violation.
Tax-exempt medical associations are highly scrutinized for antitrust violations and authorities would be quick to cite any coordinated efforts between an association and a union that may appear to have an anticompetitive effect. This could be especially problematic if the activities appeared to target certain employers. An example of such activity would be targeted boycotts of practices owned by private equity-backed firms or insurance companies.
There are many misconceptions about what unions can and cannot do. For example, physicians sometimes tell me they believe a union would negotiate better Medicare reimbursement rates. However, union negotiations are focused on the employment relationship, not reimbursement rates. In fact, to the extent I have seen union attention to Medicare, it has typically been in support of Medicare for all initiatives. Union contracts, called “collective bargaining agreements,” cover features like wages, benefits, working conditions, job protections, and time off.
Whether to form or join a union is a complicated question. Some people argue that unions help employees negotiate more effectively with employers and promote better working conditions. Others argue that unions impose rigid policies, prevent merit-based promotions, add costs, and cause conflicts with management.
There is no right or wrong answer, but as the professional landscape for physicians and surgeons evolves and more of these healthcare professionals are considering unions, it is important to gather as much information as possible to help make an informed decision.
The thoughts and opinions expressed in this column are solely those of the author and do not necessarily reflect those of the ACS.
Jeremy Lewin is a partner at Barnes & Thornburg LLP in Chicago, Illinois.